`THE SERUM CONCENTRATION OF PENICILLIN
`RABBITS, AND MEN AFTER ITS INTRAMUSCULAR
`INJECTION IN AQUEOUS SOLUTION
`
`HARRY EAGLE, RALPH FLEISCHMAN, AND A. D. MUSSELMAN
`
`Laboratory of Experimental Therapeutics of the U. S. Public Health Service and the Johns
`Hopkins School of Hygiene and Public Health, Baltimore 5, Maryland
`
`Received for publication October 13, 1948
`
`The serum levels of penicillin G in man after its intramuscular injection in
`aqueous solution or peanut-oil-beeswax suspension have been described in a pre-
`ceding paper (Tucker and Eagle, 1948). Similar data are here reported with
`respect to the plasma levels aiforded by the intramuscular injection of aqueous
`penicillin G in mice and rabbits.
`
`METHODS AND MATERIALS
`
`The solutions were injected intramuscularly into the thigh, in a volume of 1 ml
`per kg in rabbits, and in 0.2-ml volume in mice.
`In the latter species the injec-
`tion was partially subcutaneous, because of leakage from the muscle into the sur-
`rounding tissues. The courtesy of the manufacturers in supplying the penicillin
`G used in these studies (Squibb lot numbers V-31, CRA214-20, and 17579, and
`Commercial Solvents lot number 46042605) is gratefully acknowledged.
`The method of assay was a serial dilution technique in which inhibition of
`hemolysis by the C-203 strain of Streptococcus pyogenes was the end point (Ram-
`melkamp, 1942; Kirby and Rantz, 1944). The details of the method have been
`given elsewhere (Eagle and Newman, 1947).
`In the assays of human and rabbit
`serum, the data have been corrected for the inhibitory effect of serum in the assay
`(Eagle, l947a,b; Tompsett, Schultz, and McDermott, 1947a,b), using the cor-
`rective factors as experimentally determined for human serum (Eagle and
`Tucker, 1948) and since found to be valid also for rabbit serum. No corrective
`factors have been applied to the mouse data. The values in which, because of
`the high concentration of serum in the indicator tube, that correction would
`probably have been significant are indicated in the tables.
`It is a pleasure to acknowledge our indebtedness to Mr. Nathan Mantel, of the
`Oflice of Statistical Coordinator, Division of Public Health Methods, for his
`assistance in the calculation of the probable errors of the median serum levels.
`
`EXPERIMENTAL RESULTS
`
`Figure 1 and the first section of table 1, showing the results obtained in man,
`are here reproduced in modified form‘ from a previous paper (Tucker and Eagle,
`
`1 In the previous communication (Tucker and Eagle, 1948) the standard deviation given
`in the tables was a measure of the variability of the data about the median, as calculated by
`
`the formula (0 =
`
`2d’ med) and the standard error given was calculated by the for-
`n
`
`119
`
`
`
`
`
`isenfi/iq91,03‘51,Jequieidesuo/5JO'u.lSl2'C|[//Id11l.|uJoJ._1pepeo|u/v\o(]
`
`
`
`
`
`
`
`Astrazeneca Ex. 2129 p. 1
`Mylan Pharms. Inc. V. Astrazeneca AB IPR2016-01325
`
`
`
`120
`
`H. EAGLE, R. FLEISCHMAN, AND A. 1). MUSSELMAN
`
`[voL. 57
`
`1948) in order to facilitate comparison between the results obtained in the 3
`species. Figure 2 and the second section of table 1 show the median serum
`levels in rabbits after the injection of penicillin G at 60, 10, 3, 1, and 0.6 mg per
`kg. Figure 3 and the last section of table 1 give the median serum concentra-
`tions obtained in mice similarly injected.
`In general, the initial blood levels, 15 or 30 minutes after the injection, varied '
`more or less linearly with the amount of penicillin injected.
`In figure 4, the
`.»
`
`7’
`
`llnllé
`
`
`
`Penicillin G in Man
`Aqueous solution —j
`injected imramuscularly
`
`an
`
`a_._. 2-.°:6..z./:2
`
`‘
`0
`
`O
`
`\
`
`<o.os gm
`
`
`
`HEIEHHIIIIFIIIEI
` jsenfi/iq91,03‘51,Jequlejdesuo/6Jo'uJse'q_f//zdjjquJoJ._1pepeo|u/v\o(]
`
`
`
`
`
`
` O SERUMCONCENTRATIONOFPENICILLIN
`
`‘(CorrectedforSerumErrorinAssay)
`
`
`\<o.o3 7/cc
`at 4 hours
`
`Z
`3
`
`TIME IN HOURS
`
`Figure 1. The serum concentrations of penicillin G in man after its intramuscular injec-
`tion in aqueous solution at varying dosage.
`(After Tucker and Eagle, 1948; cf. footnote 1.)
`
`15-minute concentrations in man are approximations only, obtained by the
`extrapolation of the curves of figure 1. These initial values were of the same
`order of magnitude in rabbits and mice, but in man were 3 to 5 times higher.
`This is indicative of a significant species difference, either in the amount of body
`
`n
`mula S.E. = \—/‘r§—_, and is a measure of the reliability of the variability measure a. Of
`greater significance as a measure of the reliability of the median itself is its standard error,
`given by the formula S.E. med. = 1.25 - It is the latter value that is indicated
`l’ n n —
`in all the figures and tables of this paper, including the human data of figure 1 and table 1.
`
`Astrazeneca Ex. 2129 p. 2
`
`
`
`1949]
`
`SERUM CONCENTRATION or PENICILLIN G
`
`121
`
`
`
`
`
`Efifioanaousue5=o.:8...=.u.E~ao§.Ec.a=.e3.9has:“SeasE3.$.=.e§.53:gm.6=.£.8.§&He2S.£u.:§38uSatanShhuH.Hm<.H
`
`
`
`
`
`
`
`
`.2a:3:3in:2:3=2.:3~m.w.fiuW3.55flue
`
`VVV«S...Hw8.o.§.onn2...2SnEdVVH«mod§.oum$.38...um56S8m3.V«VSodHS...n8...Hmm...2...H3...2So;edVas...an39¢3...H3.mm...H“AE...anma253;3V$3.umand9.oH«NA3...Hn.»mm...Hma2So.»mwe...umgo...£5um$62....H3.”«AanS«A"H22N84:9$5
`
`...._um3$5nnH.»E.58.32anaumM8c.3umwas2..an.35.»"183HmmW2H«w3.e8638E03
`VV396H39¢36.:um.368...um$3macum$6oio8;eeVSodfl«mad8...an868...umum...mudum$556um3...9+$3HV%...oHfined3.:H2.andH356H34$5anw.NTa.8......mVmean5:.$6"1whoam...an«.4.
`
`
`
`vvv38...H:o.3s.oH:.oE...H3...“5E8;3.VVVH386H8,38...HS.o«8.oH8.9E58;HVVVEaodHm8.§8.cHasasan3nmSodmVV386um3....8...an3...3...HwdNum3.1.53.3SV38.9H368.:Hmm...3...Hm.»asHmmHwanx..38:638VEa...nm2o56H.3.54umma3umanSHSH8nm8aSumo8.mmmSn.$m=oE
`
`
`
`
`
`
`
`
`
`womannM<nz<.uman3.:nunaI<uuonufl.vzo~.n<u.nznuzoulbuumzfinflflinnc<mon
`
`
`,.w~»%.0_.%..mfiuunm
`
`
`
`MO.02
`
`
`
`
`
`._8Snmfiaumouomfimod.8no5.a.5aeo_uooAvuuoohccva3ofifinuomEon:.323wevofiufiuauzofiuam23.3nzmomnomofiasmnofiowoz_..
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`..Bu_o=_H.vnaBwanm..«3manna0&8GmEgonHOuoumvh.~Oum&_.J_Hmon?.3»vouaofioodean90¢o>.un_935Hmm_w>£nzmumnwm0&9._.dnaommawmnnoon95:.n3.3o.=mSnotEamonmoSoomoE3535.8.“mnofiuohoo€33Emw3.a>wan.HKwvmu
`
`
`
`
`
`
`
`
`
`
`
`Uo<<:_omQmQ.403::_o”sU.mm3.o6\o:mm_o._m3Uma.6.No.53ocmmfi
`
`
`
`
`
`Astrazeneca Ex. 2129 p. 3
`
`
`
`122
`
`H. EAGLE, R. FLEISCHMAN, AND A. D. MUSSELMAN
`
`[vor.. 57
`
`fluid over which the penicillin is initially distributed, or in the rate of that dis-
`tribution.
`
`In
`After the initial period of rapid absorption, the blood levels fell off rapidly.
`rabbits this rate of decrease averaged 65 per cent each hour, so that the residual
`serum concentration at time t was k X 0.35‘, in which t is the time in hours over
`the period of experimental observation, and k is a constant determined by the
`amount of penicillin injected.
`In man, the rate of fall was less consistent, the
`percentage drop per hour over the first 2 hours varying from 65 at the larger
`
`II_
`19m
`H °-01-
`§
`IILV°—
`,-—; EIIZ
`6 .
`--—_
`.. g V nIo!_
`_,
`110.4
`'a=
`lI,I—
`gs
`Iu_—
`5-: 5 EIILC
`4 5
`—°—:~,
`Es:
`.Ii\I1
`8=
`e “ii
`5 '5
`Ilg.
`° 5
`In‘!
`= l I
`g
`_
`:—
`» all
`11
`Ell
`II
`&
`//2 I
`
`2
`
`3
`HOURS
`IN
`TIME
`Figure 2. The serum concentrations of penicillin G in rabbits after its intramuscular
`injeitioln in aqueous slolutioari at vagylilng dosage. The dashed portion of the curve at 3 mg
`per g y-passes t e ow v ue at
`ours.
`
`4
`
`s
`
`U0
`
`5
`2
`8
`3
`2
`8
`3
`g
`3
`7:
`.
`8
`-3
`o
`6
`3
`3
`0)
`(D
`%3
`=7
`a»
`1
`_©
`N
`3
`c»
`0'
`
`3
`C(D
`59..
`
`In mice, however,
`doses to approximately 80 per cent at the smaller (figure 1).
`the blood levels during the first 2 hours fell oifat the extraordinary rate of 91
`8
`E
`to 99 per cent per hour (figure 3). At the lar est dosage of 200 mg per k , a
`slower rate of fall developed after several hours. This probably reflects a re-
`versal in the direction of flow of penicillin out of the tissue reservoir established
`by its initial distribution and back into the blood.
`The serum concentrations provided by a given dose of penicillin were usually
`3 to 5 times higher in man than they were in rabbits similarly injected. Al-
`though the concentrations in rabbits and mice began at the same level, they
`rapidly diverged because of the faster disappearance of penicillin in mice. These
`species diflerences are illustrated for a single dosage of penicillin in figure 5.
`
`Astrazeneca Ex. 2129 p. 4
`
`
`
`1949]
`
`snmm CONCENTRATION or PENICILLIN G
`
`123
`
`In the therapeutic use of penicillin, one of the most important single factors is
`the length of time for which the tissue levels of penicillin remain at eflectively
`bactericidal levels. When the three species were compared on this basis (figure
`6), a given serum level was sustained approximately 50 per cent longer in man
`than it was in rabbits similarly injected.
`In large part this reflects the initially
`higher concentrations provided in man.
`In turn, the levels were sustained
`
`OFPENICILLIN
`
`SERUMCONCENTRATION
`
`
`
`
`
`(No!correctedforSerumErrorinAssay)
`
`TIME
`
`IN
`
`HOURS
`
`Figure 8. The serum concentrations of penicillin G in mice after its intramuscular in
`jection in aqueous solution at varying dosage. Unlike the man and rabbit data. of figures
`1 and 2, the data in this figure have not been corrected for the inhibitory efiect of serum in
`the assay. That would robably have been significant only for the two values indicated
`by the symbol :3: in the gum.
`
`approximately 3 times longer in rabbits than they were in mice because of the
`faster rate of fall in the latter species. Thus, as is indicated in figure 6, a dose of
`0.6 mg per kg provided serum levels in excess of, e.g., 0.1 microgram per ml for
`0.5, 1.4, and 2.3 hours in mice, rabbits, and men, respectively; and after a dose
`of 10 mg per kg the corresponding time periods were 1.4, 3.9, and 6.6 hours.
`In
`order to prolong by 1 hour the time for which this level of 0.1 microgram per ml
`would be maintained in the serum, the dosage of penicillin would have to be
`approximately doubled in man, and multiplied approximately 3-fold in rabbits,
`
`
`
`
`
`isenfi/iq91,03‘51,Jequieidesuo/5J0'u.lSl2'C|_l//Id11l.|uJoJ._1pepeo|u/v\o(]
`
`
`
`
`
`
`
`Astrazeneca Ex. 2129 p. 5
`
`
`
`124
`
`H. EAGLE, R. FLEISCHMAN, AND A. D. MUSSELMAN
`
`[voL. 57
`
`whereas a 5- to 20-fold increase would be necessary in mice, depending on the
`dosage range (figure 6).
`The factors responsible for the relatively rapid disappearance of penicillin
`from mouse blood are as yet unknown. An important consideration may be the
`
`SERUMCONCENTRATIONOFPENICILLINAFTERI5MINUTES,MIOROGRAMS
`
`/"ml
`
`
`
`
`
`
`
`o.ns
`
`0.3
`
`0.6
`
`1
`
`3
`
`no
`
`0
`
`'
`
`so
`
`zoo
`
`DOSAGE OF PENIOILLIN G, M0 [K9
`Figure 4. The effect of the amount of penicillin injected intramuscularly on the initial
`(15-minute) serum concentration in man, mouse, and rabbit. Human data obtained by
`extrapolation of curves in figure 1.
`
`renal clearance of penicillin, which in this species may be higher relative to body
`weight than it is in either rabbits or man (Eagle and Newman, 1947).
`
`SUMMARY
`
`Data are given for the median serum concentrations of penicillin G after its
`intramuscular injection in aqueous solution in men, rabbits, and mice at doses of
`0.6, 1 (1.5), 3, 10, 60, and 200 mg per kg. For equal doses of penicillin the serum
`levels at a given time were in the order man > rabbit > mouse.
`The initial (15- to 30-minute) level in man was 3 to 5 times greater than it was
`
`
`
`
`
`isenfi/(q91,03‘51,Jequieidesuo/5J0'uJSl2'C|_l//Id11l.|uJoJ._1pepeo|u/v\o(]
`
`
`
`
`
`
`
`Astrazeneca Ex. 2129 p. 6
`
`
`
`
`
`SERUMCONCENTRATIONOFPENICILLING,MICROGRAMSPERml
`
`
`
`HOURSFORWHICHSERUMCONCENTRATIONREMAINED
`
`
`
`
`
`
`
`GREATERTHAN0.:mcnosrumsPERml
`
`
`
`
`
`I/4 «/2
`
`I
`
`C
`
`2
`
`
`
`
`
` isenfi/(q91,03‘51,JeC|LU81deSuo/6Jo'uJse'q[//zduquJoJ._1pepeo|u/v\o(]
`
`Figyre 5: The contrasting cnrves of the serum penicillin concentration in men, rabbits,
`and m1ce similarly injected with penicillin G.
`
`TIME
`
`IN
`
`HOURS
`
`4
`
`
`
`
`
`
`
`DOSAGE OF PENICILLIN G, MILLIGRAMS PER Kg
`
`Figure 6'. The time for which varying dosages of penicillin G_ provide serum concentra-
`tions in excess of 0.1 microgram per ml. The value for man injected at 3 mg per kg is
`an approximation only, obtained by extrapolation of the curve in figure 1.
`125
`
`Astrazeneca Ex. 2129 p. 7
`
`
`
`126
`
`H. EAGLE, R. FLEISCEMAN, AND A. n. Mussnmm
`
`[voL. 57
`
`in mice or rabbits, indicative of a significant species diflerence in the initial dis-
`tribution of the drug in the body fluids.
`Thereafter, penicillin disappeared from the serum of rabbits at the rate of 65
`per cent an hour, and somewhat faster in man.
`In mice, however, it fell off at a
`much faster initial rate, varying between 91 and 99 per cent an hour.
`In consequence of these differences in the initial distribution of penicillin and ‘
`its subsequent rate of disappearance from the serum, the time for which a given
`dosage of penicillin provided a measurable serum concentration was in the order
`1 :3 :4% in mouse, rabbit, and man, respectively.
`To prolong a given serum level by 1 hour would require approximately a 2-fold
`increase in dosage in man, a 3-fold increase in rabbit, but a 5- to 20-fold increase
`in mice.
`
`REFERENCES
`
`1947a The inactivation of penicillins F, G, K, and X by human and rabbit
`EAGLE, H.
`serum. J. Exptl. Med., 85, 141-161.
`EAGLE, H.
`1947b The varying blood levels aflorded by penicillins F, G, K and X in rabbits
`and man. J. Exptl. Med., 86, 163-173.
`EAGLE, H., AND NEWMAN, E. V.
`1947 The renal clearance of penicillins F, G, K and X in
`rabbits and man.
`J. Clin. Investigation, 28, 903-918.
`EAGLE, H., AND Tucxmn, H. A.
`1948 The efiect of human serum on the dilution bioasssy
`of penicillins G, X, and K.
`J. Bact., 56, 59-62.
`Kmnr, W. M., AND RAN-rz, L. A.
`1944 Methods of measuring penicillin concentrations
`in body fluids.
`J. Bact., 48, 603.
`RAmmLxAm>, C. H.
`1942 A method for determining the concentration of penicillin in
`body fluids and exudates. Proc. Soc. Exptl. Biol. Med., 51, 95.
`Toupsm-r, R., Scmmrz, 8., AND McDnnno'r-r, W.
`1947a The relation of protein-binding
`to the pharmacology and antibacterial activity of penicillins X, G, dihydro F, and K.
`J. Bact., 53, 581-595.
`Influence of protein-binding
`1947b
`Toursnrr, R., Scnur/rz, S., AND McDano*n', W.
`on the interpetation of penicillin activity in viva. Proc. Soc. Exptl. Biol. Med., 65.
`163-172.
`1948 The serum concentrations of penicillin G in man
`Tucxnn, H. A., AND EAGLE, H.
`following its intramuscular injection in aqueous solution and in peanut oil-beeswax
`suspension. Am. J. Med., 4, 343-354.
`
`
`
`
`
`isenfi/iq91,03‘51,Jequieidesuo/5J0'u.lSl2'C|_l//Id1J,l.|uJoJ._1pepeo|u/v\o(]
`
`
`
`
`
`
`
`Astrazeneca Ex. 2129 p. 8